A model for assessing the relative humaneness of pest animal control methods. Second edition June 2011

(1)

A model for assessing the

relative humaneness of pest

animal control methods

Second edition June 2011

by Trudy Sharp and Glen Saunders

(2)

A model for assessing the relative humaneness of pest animal control methods. Second edition 2011 ISBN: 978-1-921575-26-6

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Commonwealth. Requests and inquiries concerning reproduction and rights should be addressed to the Commonwealth Copyright Administration, Attorney General’s Department, Robert Garran Office, National Circuit, Barton ACT 2600 or posted at http://www.ag.gov.au/cca Disclaimer:

The Australian Government Department of Agriculture, Fisheries and Forestry seeks to publish its work to the highest professional standards. However, it cannot accept responsibility for any consequences arising from the use of information herein. Readers should rely on their own skill and judgment in applying any information for analysis to particular issues or circumstances.

Cover photos:

Feral cat (Felis catus) (photo by Robert Brandle)

Cane toad (Bufo marinus) (photo by Kimberley Toad Busters) Wild dog (Canis sp.) (photo by Peter Fleming)

Rabbit (Oryctolagus cuniculus) (photo by NSW Department of Primary Industries) Citation:

This document should be cited as:

Sharp, T. and Saunders, G. (2011). A model for assessing the relative humaness of pest animal control methods (Second edition). Australian Government Department of Agriculture, Fisheries and Forestry, Canberra, ACT. Printed by: New Millennium Print

(3)

Preface to the Second Edition

This second edition has been updated and revised to include minor modifications and improvements to the model for assessing the relative humaneness of pest animal control methods. Some of the examples in the impact scales have been modified and notes have been added to the worksheet to clarify some aspects of the assessment process.

This edition also contains a summary of the project which applied the model to commonly used control methods. Completed assessments for a range of species and techniques are included as well as ‘humaneness matrices’ which provide a simplified overview of the relative humaneness of all the methods for each species.

Thus, Section 1 contains a review of

humaneness assessment and a summary of the project to develop the model whilst Section 2 describes the process undertaken to assess the humaneness of commonly used invasive animal techniques.

Information in this publication has been

summarised below to provide a quick overview of the process to develop and apply the model and we have also addressed some of the issues encountered along the way:

Background to the Assessment

model

A workshop hosted by RSPCA Australia, the Animal Welfare Science Centre and the

Vertebrate Pests Committee (VPC) was held in 2003 in Melbourne. It was attended by

representatives from government (State and Commonwealth), animal welfare and veterinary organisations e.g. Australian Veterinary

Association (AVA), RSPCA and pest animal control organisations e.g. Livestock Health and

Pest Authorities as well as producer bodies e.g. Cattle Council, Australian Wool Innovation, Victorian Farmers Federation.

The workshop examined solutions for achieving humane pest animal control and identified a major weakness in the consideration of animal welfare. While the workshop participants indicated there was a will to include consideration of animal welfare in control strategies and in the registration of new control products, a process to do this in an objective, science based way was lacking. This lack of animal welfare consideration was viewed as posing a general threat to ongoing pest animal control operations.

With the financial support of the Department of Agriculture, Fisheries and Forestry under the Australian Animal Welfare Strategy (AAWS), a project to develop a process for assessing the relative humaneness of pest animal control methods was offered to tender in 2007. Under the management of a steering group formed from members of the AAWS Wild Animals Working Group, the NSW (I&I) Vertebrate Pest Research Unit was commissioned to develop a suitable model for humaneness assessment.

Development of an Assessment

Model

Creating a suitable, workable model proved to be difficult due to the variety of control

techniques used, the wide range of pest animals targeted, and the inclusion of both lethal and non-lethal methods. The final aim therefore became to produce a practical, general model of humaneness assessment that can be applied to any pest control method.

The assessment of overall welfare impact is based on five domains:

(4)

1: Thirst/hunger/malnutrition

2: Environmental challenge

3: Injury/disease/functional impairment 4: Behavioural/interactive restriction

5: Anxiety/fear/pain/distress

The model was not designed to provide an absolute measure of humaneness but will allow a judgement to be made about the impact of a specific control method on the target animal. When the model is applied to a range of different methods, these can be compared and a decision can be made on the choice of method that is informed by an understanding of the relative humaneness of each method being considered.

It was clear from the initial tender that the development of an assessment process would require significant stakeholder input and agreement for it to have a wide uptake and ultimate impact. As part of the consultation process a workshop to consider a first draft of the humaneness model was held in April 2008 and was again attended by representatives from various State/Territory and Commonwealth governments (including the CSIRO and APVMA), and non-governmental organizations such as RSPCA, Animals Australia, NSW Farmers and AVA. At the workshop it was agreed that the model was acceptable and would be workable with some minor modifications. The modified model was published in 2008.

Discussions on how the model should be used were also undertaken at the workshop with the majority of stakeholders expressing their support for its application to currently used control methods (those for which Standard Operating Procedures had been written). There was general agreement that a panel of experts with knowledge and experience in animal welfare and invasive animal management should determine the priority methods for assessment, complete the assessments using the

humaneness model and then disseminate the results to a wider audience. It is anticipated that the information gained will improve best practice management of invasive animal species by enabling humaneness to be considered alongside efficacy, cost-effectiveness, practicality, target specificity, operator safety etc. when determining the most appropriate method for managing the impact of an invasive animal.

The Assessment Process

In 2009, a project was funded by the Australian Pest Animal Research Program (APARP) formerly known as the Australian Pest Animal Management Program (APAMP) and the Australian Animal Welfare Strategy (AAWS) through the Commonwealth Department of Agriculture, Fisheries and Forestry (DAFF) to identify and coordinate a suitable panel, who would then apply the model to a selection of currently used invasive animal methods.

The objectives of the Humaneness Assessment Panel were to:

• Produce a priority list of methods to be assessed.

• Identify other suitable advisors with expertise and experience in particular invasive animal species and their

management. These ‘species experts’ will be invited to attend meetings of the panel and participate in the humaneness assessments for their species of interest.

• Review current information and perform humaneness assessments using the model as a framework.

• Identify where there is a need to develop more humane methods or to expedite the introduction of more humane methods. • Identify areas where there are gaps in

knowledge regarding the welfare impact of control methods. These gaps will be

(5)

discussed in the final project report and also reported to the AAWS research group and the VPC Animal Welfare technical group.

• Report any problems, suggested changes or other issues with the current Standard Operating Procedures for the Humane Control of Pest Animals to the Vertebrate Pest Research Unit, NSW Dept. Industry and Investment so that modifications can be made when the SOPs are reviewed. • Provide suggestions, where appropriate, for

improving the assessment model.

• Release the results of the humaneness assessments so they can be placed on a suitable website e.g. feral.org.au or DAFF AAWS (website yet to be established).

• Seek endorsement of the assessments by the National Vertebrate Pest Committee.

• At the end of the project’s tenure, make recommendations on the requirement for, and composition of future panels to conduct further assessments and/or to periodically review completed assessments to take into account new research.

The Humaneness Assessment Panel was based on expertise and independence and comprised of:

Dr Glen Saunders Research Leader

Vertebrate Pest Research Unit, Industry and Investment NSW

Dr Bidda Jones Chief Scientist RSPCA Australia Mr Chris Lane

Program Coordinator, Invasive Animals Cooperative Research Centre

Mr Jason Neville

Senior Ranger, Pest Management, Western Rivers Region

DECC - Parks and Wildlife Group Ms Trudy Sharp

Project Officer

Dr Andrew Fisher

Associate Professor in Production Animal Management and Welfare

Faculty of Veterinary Science, University of Melbourne

Dr Frank Keenan

A/Manager, Land Protection Policy Biosecurity Queensland

Dr Andrew Braid Veterinarian

CSIRO Sustainable Ecosystems, Agricultural Landscapes Program

Additional members were co-opted for their specific species or technique expertise. No techniques were assessed without this additional expertise being available.

Further information on the panel members and invited species experts are included in Section 2. The Assessment Panel’s proceedings were recorded and reported to AAWS and APAMP in the project’s final reports.

Issues

Some concerns were expressed over the implementation of the process for assessing the relative humaneness of pest animal control methods, under what circumstances it might be applied and how it might threaten the use of existing pest animal control strategies. In relation to some of these concerns the following points are relevant:

• There is a worldwide trend towards ethical and moral concern for welfare of animals

(6)

regardless of their status. This trend cannot be ignored.

• The ranking process is an enabler not an inhibitor; it provides support to control techniques through evidence based, objective assessment.

• When selecting the most appropriate pest control techniques to apply in the field, welfare is just one of the issues to be considered. This is made clear in the documentation around the model. • Control techniques are relatively easy to

assess on the basis of efficacy, cost/benefit, target specificity etc. but nothing was previously in place to assess humaneness. • The development of a humaneness

assessment for a particular control

technique is not linked to a decision on the need to implement control; this is a given.

• Application of the model provides transparency to the community and is independent of the end-users which is to their advantage.

• The use of the model and the assessment results provides a means of encouraging development of more humane techniques.

• The consultation process involved in this process was extensive. No groups were deliberately excluded.

Glen Saunders and Trudy Sharp June 2011

(7)

Section 1 - Developing a model for assessing the

relative humaneness of pest animal control methods

(10)

(11)

Foreword

Consideration of animal welfare in the management of invasive animals is essential to ensure that control techniques are performed humanely. For this reason, a workshop hosted by RSPCA Australia, the Animal Welfare Science Centre and the Vertebrate Pests Committee was held in 2003 in Melbourne, Australia. The workshop examined solutions for achieving humane invasive animal control and identified a major stumbling block in the consideration of animal welfare. While the workshop participants indicated there was a will to include animal welfare in control strategies and in the registration of new control products, what was lacking was an accepted way to do this. In other words, to properly consider humaneness in invasive animal management, we needed to have a reliable and practical method of assessing it.

After further thought and discussion, and with the financial support of the Department of Agriculture, Fisheries and Forestry under the Australian Animal Welfare Strategy (AAWS), a project to develop a model for assessing the relative humaneness of pest animal control methods commenced in April 2007. Under the management of a steering group formed from members of the AAWS Wild Animals Working Group, Trudy Sharp and Glen Saunders, from the NSW Department of Primary Industries Vertebrate Pest Research Unit, were commissioned to develop the model. It was clear that the model would require significant stakeholder input and agreement for it to have any chance of a wide uptake and ultimate impact. The project included broad consultation and the direct involvement of a range of stakeholders, with the goal of achieving the eventual endorsement of those individuals and groups.

Creating a suitable, workable model proved to be a difficult process due to the variety of control techniques used, the wide range of pest animals targeted, and the inclusion of both lethal and non-lethal methods. The final aim therefore was to produce a practical, general model of assessment that can be applied to any pest control method.

The model does not give an absolute measure of humaneness: it is designed to allow a judgement to be made about the impact of a specific control method on the target animal. When the model is applied to a range of different methods, these can be compared and a decision can be made on the choice of method that is informed by an understanding of the relative humaneness of each method being considered.

The model presented here provides a reliable, functional and accepted method that enables humaneness to be considered as an integral part of planning invasive animal control. The next step is for those involved in the decision making process, including government agencies, registration authorities and land managers, to ensure its uptake and application. I commend the model to you.

Bidda Jones

Leader, Project Steering Committee Chief Scientist, RSPCA Australia 2008

(12)

Acknowledgements

We would like to acknowledge the input and guidance of the project steering group with special thanks to Dr Bidda Jones and Ms Kristy McPhillips for their invaluable support and assistance. We would also like to thank the numerous stakeholders who provided useful comments throughout the project and also Professor David Mellor for his suggestions on improving the humaneness model. This project has been funded by the Commonwealth Department of Agriculture, Fisheries and Forestry (DAFF) under the Australian Animal Welfare Strategy (AAWS).

(13)

Executive Summary

Pest animals such as rabbits, feral pigs, foxes, wild dogs and feral cats continue to cause significant environmental damage and agricultural losses in Australia despite improvements in control methods and the development of new techniques. Each year hundreds of thousands of pest animals are trapped, poisoned, shot or otherwise destroyed because of the harm they cause (Olsen 1998). Historically, pest animal control has focused on killing as many pests as cheaply as possible. For most people in today’s society the management of pest animals is acceptable provided that such management is humane (Mellor and Littin 2004) and justified. However, many of the methods used to control pest animals in Australia are far from being humane. There is a pressing need to improve the humaneness of control programs and to develop a process that enables the most humane methods to be identified.

The ‘humaneness’ of a pest animal control method refers to the overall welfare impact that the method has on an individual animal. A relatively more humane method will have less impact than a relatively less humane method. The development of a system to assess the relative humaneness of control techniques was identified as a priority at a joint workshop held by RSPCA Australia, the Animal Welfare Science Centre and the Vertebrate Pests Committee in 2003 (Humane Vertebrate Pest Control Working Group 2004). Information from such a system could be used to assist decision makers in the development, planning and implementation stages of pest animal control programs along with other factors such as efficacy, cost-effectiveness, practicality, target specificity and operator safety.

Included in this report is a review of current information relating to the assessment of humaneness and welfare impact. It examines the assessment of welfare in laboratory animals, production animals and wild animals and also

summarises methods used to determine the welfare impact of some pest animal control methods. Based on this review, it was apparent that although there are some systems for assessing humaneness for specific classes of control methods (i.e. injury scoring for restraining traps, comparison of poisons), there are none that could be applied to the full range of pest animal control techniques used in Australia. A model was therefore developed to achieve this aim.

The model presented in this report examines the negative impacts that a control method has on an animal’s welfare and, if a lethal method,

how the animal is killed. There are two parts:

Part A examines the impact of a method on overall welfare and the duration of this impact; Part B examines the intensity of suffering and duration of suffering of the killing technique. In Part A, overall welfare impact is assessed by looking at the impact in each of five ‘domains’, originally described by Mellor and Reid (1994) to examine the impact of scientific procedures on experimental animals. Domain 1 is water deprivation, food deprivation and malnutrition; Domain 2 is environmental challenge; Domain 3 is injury, disease, functional impairment; Domain 4 is behavioural, interactive restriction; and Domain 5 is anxiety, fear, pain and distress. The degree of impact in each domain is rated on a five-step scale – no impact, mild, moderate, severe or extreme impact. The overall impact is the rating given to Domain 5 since this represents the outcome of the impacts in the other four domains (and also includes external influences, such as the presence of humans). In Part B, the killing method is assessed by examining the level of suffering and the duration of suffering based on the time to insensibility base on the criteria described by Broom (1999). Matrices are used to determine the score for each part and then the two scores are combined to obtain the overall humaneness score.

(14)

The main advantage of this model is that it provides a systematic, comprehensive and transparent process that helps to generate consensus among diverse stakeholders regarding the humaneness of control methods. Also, the

relative humaneness of different techniques

can be compared based on the score obtained. Although it cannot achieve a purely objective and precise assessment, this model allows us to

grade humaneness using the available scientific

information and informed judgement

The humaneness model has received widespread support with the majority of stakeholders indicating that it is effective and practical. Consultation with relevant stakeholders has also indicated that they support the application of the model by an expert panel to currently used control techniques and that the assessments be disseminated to a wider audience.

Membership of the

project steering group

The project steering group included representatives from the AAWS Animals in the Wild Working Group:

■ Bidda Jones, RSPCA Australia

■ Chris Buller, Invasive Animals CRC

■ Frank Keenan, Department of Primary

Industries and Fisheries, Queensland

■ Maxine Cooper, Australian Capital Territory

(ACT) Government

■ Kristy McPhillips, Australian Department of

Agriculture, Fisheries and Forestry (DAFF)

■ Tony Peacock, Invasive Animals CRC ■ Quentin Hart, Bureau of Rural Sciences

And also:

■ Kate Littin, New Zealand Ministry

of Agriculture and Forestry

Scope and approach

The objectives of this project were to:

Phase 1:

■ Undertake a desktop review and evaluation

of existing literature (i.e. studies, articles, documents, codes of practice, standard operating procedures etc.) relating to the assessment of humaneness of pest animal control methods. Information obtained from this review will be used to develop a humaneness ranking model that contains key welfare assessment principles. The purpose of the model is to allow the relative humaneness of control methods to be taken into consideration during the development, planning and implementation stages of pest animal control programs.

■ Submit the draft model via a scoping

document to the steering group for approval.

Phase 2:

■ With the assistance of the steering group,

identify key stakeholders with an interest or involvement in the use of pest animal control techniques (i.e. APVMA; farmers; animal welfare organisations; land managers - government and non-government; and the community).

■ Circulate the scoping document to

identified stakeholders to obtain feedback.

■ Collate all comments received from

stakeholders and incorporate these comments into a new draft of the scoping document. Submit new draft of scoping document to the steering group for approval.

(15)

Phase 3:

■ Circulate the second draft of the scoping

document to stakeholders.

■ Identify main points of agreement/difference

and prepare agenda for stakeholder meeting based on these points.

■ Organise a face-to-face meeting of key

stakeholders with the aim to reach consensus on the proposed ranking of humaneness model.

■ Prepare a report of the meeting and

prepare a final version of the ranking of humaneness model. Submit final report to steering group, who will submit to high level stakeholders e.g. National Resource Management Ministerial Council (NRMMC), Primary Industries Ministerial Council (PIMC), Australian Pesticides and Veterinary Medicines Authority (APVMA) for endorsement.

The outcomes of the project were:

■ A teleconference involving members of the

steering group and consultant was held on 31 May, 2007. The aim was to discuss the proposed approach and to receive some initial feed back on the consultancy.

■ A desktop review of literature relating to

the assessment of humaneness and a draft humaneness ranking model was prepared and circulated to members of the steering group on 10 August, 2007. Comments and suggestions were incorporated into a second draft and this was circulated to stakeholders with an interest or involvement in pest animal control on 22 November, 2007

■ A discussion paper was prepared which

included a summary of the 36 comments received from stakeholders and also formed the basis of the agenda for a workshop to discuss and refine the proposed humaneness model.

■ A workshop to discuss the proposed model

for assessing the relative humaneness of pest animal control methods was held on Wednesday 9th April, 2008. Twenty-seven invited participants, including members of the project steering group, attended the workshop including representatives from State/Territory and Commonwealth governments (except ACT) as well as the CSIRO, Australian Pesticides and Veterinary Medicine Authority (APVMA), Animals Australia, NSW Farmers, Australian Veterinary

Association and Massey University, New Zealand. A report summarising the outcomes of the workshop was prepared and comments and suggestions were incorporated into a final version of the humaneness model.

■ A final report that incorporates the

literature review and the model for assessing humaneness was submitted.

(16)

(17)

Part A: Review of humaneness assessment

pest management, but only recently has there been an interest in assessing the impact that control methods have on animal welfare (see Littin and O’Connor 2002; Mason and Littin 2003; Morris et al. 2003; O’Connor et al. 2003; Jones 2003a; Littin 2004; Mellor and Littin 2004; Littin and Mellor 2005). A commonly held view in today’s society is that the use and management of animals by humans is acceptable provided that such use and management is humane (Mellor and Littin 2004) and justified. This review will firstly define the concept of humaneness and how it relates to animal welfare. It will then summarise some of the current approaches to the assessment of humaneness/welfare in a range of different animal types and specific situations.

A2. What is humaneness?

A 2a Defining humaneness

To assess humaneness objectively we need to define it; this is not an easy task. Most dictionaries classify the word ‘humane’ as an adjective that describes a particularly human quality e.g. ‘marked by compassion, sympathy or consideration for human or animals’ (Merriam-Webster’s Collegiate dictionary); ‘having or showing compassion or benevolence’ (Oxford Dictionary); and ‘having a disposition to treat other human beings or animals with kindness’ (Webster’s Dictionary). Yet in documents relating to the treatment of animals, the word humane is used to mean causing minimum pain and suffering, most often in relation to killing methods. For example, the

A1. Introduction

Pest animals continue to be a significant problem in Australia despite improvements in pest animal control methods and the development of new techniques. Each year hundreds of thousands of foxes, rabbits, kangaroos, goats, pigs, mice, cats, rats and birds are trapped, poisoned, shot or otherwise destroyed because of the agricultural losses and the environmental harm they cause (Olsen 1998). Methods used in the management of pest animals include:

■ lethal methods such as shooting, poisoning,

gassing, introduction or encouragement of specific disease, capturing an animal using a trap, snare or net and then killing it, destruction of burrows containing animals using explosives or ripping; and

■ non-lethal methods such as exclusion fencing,

repellents and deterrents, fertility control, harbour removal, live capture and release of animal (Sharp and Saunders 2005).

The main aim of pest management is not to kill large numbers of pest animals but to reduce pest animal damage and to promote sustainable production and/or the conservation of biodiversity (Olsen 1998). To achieve this aim, a strategic approach to pest animal management is recommended (Braysher 1993). This involves the use of scientifically based procedures that are humane, cost-effective and integrated with ecologically sustainable land management. Over the years there has been much research looking at the economic and ecological elements of

(18)

RSPCA Australia policy on humane killing states that to achieve a humane death ‘an animal must be killed instantly or instantaneously rendered insensible to pain until death supervenes’ (Jones 2003b). Broom (1999) uses the term ‘humane killing’ to refer to instances where the welfare of the animal is not poor just prior to the initiation of the killing procedure and the procedure itself results in insensibility to pain and distress within a few seconds. Jones (p9, 2003b) describes both aspects of the word when she states that “the humaneness of a killing method can be measured either by the absence of pain, suffering or distress experienced by the animal, or by the relative level of compassion and kindness exhibited by humans”.

The term humane can be confusing and it often attracts controversy whenever it is used. In 1997, The International Standards Organisation process to adopt internationally agreed humane trapping standards was stopped because an agreement on the definition of the term humane could not be reached (Harrop 1998). Rather, it was agreed to work on ‘trap testing methodology standards’ instead of ‘humane trapping standards’. In 1999, The International Whaling Commission (IWC) removed the word humane from the title “Working Group on Welfare Considerations of Whale Killing Methods” and the related “Workshop of Whale Killing Methods” because a number of countries objected to the very subjective nature of the term and its failure to reflect differences in cultural and traditional backgrounds (Gillespie 2003). In the Codes of Practice for Humane Pest Animal Control (Sharp and Saunders 2005)

humane was defined as:

“causing the minimum pain, suffering and distress possible. To be humane is to show consideration, empathy and sympathy for an animal, an avoidance of (unnecessary) stress, and the demonstration of compassion and tenderness towards our fellow creatures” (Australian Veterinary Association 1997).

In a recent review by stakeholders, a decision was made to delete this definition as it was considered to be an inappropriate starting

point for defining methods of pest animal control (Braid and Buller 2007).

A definition of humane that may be more relevant to pest animal control is “a desire to avoid the infliction of unnecessary pain upon wild animals”(Gillespie 2003). As such, when animals are to be legitimately killed, it must be done in a way that causes minimum pain and reduces the time to death wherever possible. Humane vertebrate pest control (HVPC) has been defined as “the development and selection of feasible control programs and techniques that avoid or minimise pain, suffering and distress to target and non-target animals” (Humane Vertebrate Pest Control Working Group 2004). A totally humane pest animal control method would therefore not cause any pain, suffering or distress.

Therefore, in the case of pest animal control, humaneness should not only refer to a killing method but should also extend to what happens to the animal prior to killing or to the effects of non-lethal methods used for pest animal control (e.g. live traps, exclusion fencing, deterrents). When we talk of the ‘humaneness’ of a control method, we are really talking about the overall impact that a control method has on an individual animal, and when we talk about impact, we really mean the impact on that animal’s welfare. A relatively more humane method will therefore have less negative impact on an animal’s welfare than a relatively less humane method. There is no one pest control method that does not have some sort of impact on an animal, therefore to compare humaneness of methods we have to compare these impacts.

A 2b Why do we need to assess

the relative humaneness of

pest animal control methods?

Pest animal control operations can cause a range of negative impacts on both target and non-target animals, resulting in harm and suffering. To reduce animal suffering, the most humane methods that are useable in any given situation must be employed. In order to use the most

(19)

humane control method, we need to be able to evaluate the humaneness of a technique. To assess humaneness, we need to assess what harms are being done to an animal and how bad each harm is with respect to intensity and duration (Mellor and Littin 2004). The concept of ‘relative humaneness’ refers to the degree to which a technique causes pain, suffering or distress. Evaluating which methods are more or less humane enables us to choose the most humane method for a particular situation. If we are to choose the method that causes the least suffering and distress, it is essential that we are at least able to recognise adverse effects and in some cases be able to quantify these effects.

A3. Defining animal welfare

The term ‘animal welfare’ is often used in scientific literature, legislation, public statements and general discussion. However, the concept of animal welfare is often difficult to define and is subject to continuing debate. Dawkins (2006) states that “good animal welfare” involves physical health and positive emotions, such as pleasure and contentment. “Poor welfare” comes not only from ill-health, injury and disease but also from negative emotions such as frustration or fear, which we call suffering. Broom (1996) states that the welfare of an individual is its state as regards its attempts to cope with its environment. It is a characteristic of an animal, not something given by humans and it will vary on a continuum

from very good to very poor. He argues that welfare should be defined in such a way that it can be readily related to other concepts such as: needs, freedoms, happiness, coping, control, predictability, feelings, suffering, pain, anxiety, fear, boredom, stress and health. Scott et al. (2003) define welfare as a complex construct that combines both subjective and objective aspects of the conditions of life for animals. Fraser (1993) prefers to use the term ‘well-being’ to refer to the state of the animal and uses ‘animal welfare’ to refer to the broader concept that includes social and ethical issues. In this review, the term ‘animal welfare’ will allude to a complex construct that includes both objective and subjective aspects of the physical and mental well-being of animals.

A4. Assessment of animal

welfare

A 4a How is welfare assessed?

A key issue in the assessment of welfare is that it should consider what matters to animals from their point of view (Bracke et al. 2002). The general methods for assessing welfare involve the use of:

■ direct indicators of poor welfare;

■ tests of avoidance;

■ tests of positive preference;

(20)

■ measures of ability to carry out normal

behaviour and other biological functions; and

■ direct indicators of good behaviour

(Broom 2007).

A large number of objective measures of welfare can be used in an attempt to determine the welfare state of an animal. A summary of these are included in Box 1:

Box 1: Measures of welfare (from Broom 2007)

■ physiological and behavioural

indicators of pleasure;

■ extent to which strongly preferred

behaviours can be shown;

■ variety of normal behaviours shown

or suppressed;

■ extent to which normal physiological

processes and anatomical development are possible;

■ extent of behavioural aversion shown; ■ physiological and behavioural attempts

to cope;

■ immunosuppression;

■ disease and body damage prevalence; ■ behaviour pathology;

■ brain changes;

■ body damage prevalence;

■ reduced ability to grow or breed; and ■ reduced life expectancy.

Although there exists a multitude of different welfare measures it is generally agreed that there is no one single measure or standard welfare ‘thermometer’ that can be used by itself to tell us the state of an animal (Mason and Mendl 1993; Bracke et al. 1999a; Dawkins 2004). Therefore,

a number of indicators from a variety of areas (i.e. health, physiology and behaviour) are required to get an overall picture of an animal’s welfare. A common strategy for assessing welfare involves constructing lists of the most important welfare indicators as determined by consensus of expert opinion (e.g. Whay et al. 2003; Rousing et al. 2007). Some assessment protocols also use a weighting process with the most important indicators attracting a higher weight. An overall welfare score is obtained by summing the weighted scores for each of the indicators (Bracke et al. 1999a; Bracke et al. 2002; Bracke 2006).

Another approach to assessment of welfare relies more heavily on behavioural observations to capture both the physical and mental aspects of welfare. Dawkins (2004) argues that, instead of constructing lists of many different welfare indicators, welfare assessment should be directed at answering two key questions: (1) Are the

animals healthy?; and, (2) Do they have what they want? Answers to these questions summarise

what most people need to know about animal welfare and guide the process of collecting the most relevant evidence. Observing an animal’s behaviour can be a less intrusive way of assessing welfare and avoids some of the difficulties associated with the interpretation of physiological parameters.

A more subjective approach to assessing welfare is to evaluate an animals ‘Quality of Life’ (QoL) (see Scott et al. 2003; Broom 2007; Kirkwood 2007; Scott et al. 2007). QoL has been defined as:

“the subjective and dynamic evaluation by

the individual of its circumstances (internal and external) and the extent to which these meet its expectations (that may be innate or learned and that may or may not include anticipation of future events), which results in, or includes, an affective (emotional) response to those circumstances (the evaluation may be a conscious or an unconscious process, with a complexity appropriate to the cognitive capacity of the individual)”(Scott et al. 2007).

(21)

Some argue that QoL is essentially the same as welfare, the difference being that welfare is considered over the short-term or long-term, whereas QoL refers to the characteristic of an individual over a time-scale longer than a few days (Broom 2007). A QoL approach has been used to develop a number of health-related quality of life instruments to assess acute and chronic pain in dogs and has also been generalised to farm animal welfare (Scott et al. 2003).

One of the main problems associated with the assessment of welfare is that our interpretation of the many objective welfare measures involves subjective judgements which are in turn influenced by the nature and extent of our concern for the animal under consideration (Mason and Mendl 1993). Also, although the mental state of an animal is an important aspect of its welfare; recognising and assessing this is far from easy. Measurement of animal welfare is always going to be a difficult process. Although we have a range of objective physiological and behavioural changes that can indicate poor welfare, these measures can be difficult to interpret. It can sometimes be very difficult to know if an animal is suffering because we do not have access to its state of mind and so do not know what it is actually feeling. What we can do though is scientifically collect evidence from which we can make inferences about its welfare state (much like a doctor who uses signs and symptoms to make a judgement about a disease) (Mason and Mendl 1993).

A 4b Assessment of laboratory

animal welfare

A major concern about the use of animals in research and testing is the potential for scientific procedures to cause pain, suffering or distress (Hawkins 2002). The ‘three R’s’ concept of replacement, reduction and refinement, first proposed by Russell and Birch in their book, ‘Principles of Humane Animal Experimentation’ have been incorporated into the national legislation of many countries and have become widely

accepted by the scientific community (Buchanan-Smith et al. 2005). Replacement involves using non-animal alternatives where available, whilst reduction involves reducing the number of animals used for procedures. Refinement of scientific procedures involves minimising any pain or suffering that might be experienced by animals. To assist with achieving refinement, a number of techniques have been developed for animal monitoring and to aid the recognition of discomfort, pain and distress. These include score sheets (e.g. Mertens and Rulicke 1999; van der Meer et al. 2001), clinical observation sheets, severity scales (Mellor and Reid 1994) and harm-benefit analysis.

A survey of scientific establishments was recently undertaken in the UK to evaluate how pain, suffering and distress are recognised in laboratory animals (Hawkins 2002; 2003). It was found that

clinical observation sheets are widely used to note

simple objective measures such as body weight and for logging inspection times and any observed adverse effects. Also used are score sheets which were originally suggested by Morton and Griffiths (1985). The principle behind score sheets is that observations of clinical signs are used as a way of determining the degree to which an animal’s physiology and mental state has deviated from normal, and then using these changes to make an assessment of the severity of the adverse effects (Morton 1998). It is assumed that those making the assessment will have a good knowledge of the animal’s normal behaviour and physiology. Score sheets are usually made up specifically for each scientific procedure and for each species. They list key clinical signs and behaviours that are associated with discomfort, pain and distress along with objective measures of health and/or development such as body weight. These criteria are assigned numerical scores so that an overall or total score can be produced that represents the overall adverse welfare effects. More recently the score sheet has evolved to use binary scoring, whereby the clinical signs are marked as simply present or absent, rather than using numerical scores. Other techniques for assessing animal

(22)

well-being and recording observations included data management systems, phenotype assessment protocols and visual analogue scales.

Hawkins (2002) points out that the main problem with the assessment of laboratory animal welfare is that it is still largely a subjective exercise. There are few, if any, specific, objective behavioural indicators of pain, suffering and distress and the systems that are currently in use are heavily reliant on subjective criteria. The author concludes that binary score sheets appear to be the most effective way of assessing animals and recording observations and can be a useful tool for improving objectivity and consistency in many situations.

Harm-benefit analysis is a major feature of the ethical review that animal ethics committees undertake when they consider applications to conduct research, teaching and testing procedures on live animals (Mellor 2004). The harm-benefit analysis examines the balance between the expected severity of the welfare compromise and the expected benefits of the procedure. To assist in the comprehensive assessment of the harms caused by scientific procedures, Mellor and Reid (1994) have developed a severity scale based on the UK Farm Animal Welfare Council’s ‘Five Freedoms’. This approach is based on the notion that an animal’s welfare is good when its nutritional, environmental, health, behavioural and mental needs are met. The following five domains of potential animal welfare are identified:

■ Domain 1 is food deprivation/water

deprivation/malnutrition1_;

■ Domain 2 is environmental challenge; ■ Domain 3 is disease/injury/functional

impairment;

■ Domain 4 is behavioural restriction; and ■ Domain 5 is anxiety/fear/distress.

1_{This domain was originally named thirst/hunger/malnutrition} but was re-named after realising that thirst and hunger should properly be located in Domain 5

Research proposals are examined systematically in all domains, and the degree of welfare compromise in each is rated on a 5-step non-numerical scale (O, A, B, C, X). Anxiety/fear/pain/distress arising from compromise in domain 1-4 is cumulated in to domain 5. The overall rating is commonly that given to domain 5, but if the score for this domain is low or unknown, it is given to the highest rating in the other domains. The major advantage of this system for assessing the impact on welfare is that it encourages systematic consideration of all sources of possible compromise (Bayvel 2000). This wider consideration allows more accurate assessment of the severity of impact and thereby improves the validity and efficiency of a harm-benefit analysis. Another advantage is that it predicts welfare compromise in advance and therefore can prevent it. Concerns have been raised, however, about the potential for a lack of consistency in the way the scale is applied. Because qualitative terms such as mild, moderate, short-term etc. are used in the grading system, any assessment or prediction of impact will require a subjective judgement of what these terms actually mean in a specific situation. It has also been suggested that the purpose of the scale is not well understood by some people using it with the result being that the category descriptors and examples are seen as prescriptive requirements rather than the guidelines they were intended to be (Mellor et al. 2005). When assessing individual cases, the authors of the scale have stressed the importance of applying a degree of judgement when determining the anticipated impact. The categories and guidelines are meant to be flexible and should not be seen as definitive or precise descriptors of impacts (Mellor and Reid 1994).

The severity scale developed by Mellor and Reid has been used in New Zealand since 1997 to assess and record the level of animal welfare compromise imposed by research, testing and teaching. The data from these assessments are required by law to be submitted to the NZ Ministry of Agriculture and Forestry in an annual return. Recently, a review was undertaken to

(23)

examine the operation and effectiveness of the scale and the extent to which it fulfils the purpose for which it was devised (Mellor et al. 2005). Recommended revisions outlined in the review included the following:

■ the name of the categorisation system

should be the “impact scale” rather than the “severity scale”. The word impact should replace the words severity and suffering to acknowledge that while there will always be an impact, suffering does not always occur;

■ the current 5-step non–numerical scale

should be enlarged to include a sixth category (labelled Z) which includes procedures that should not be carried out under any circumstances;

■ an exhaustive list of manipulations with

recommended gradings is not advisable, because it will inevitably not be comprehensive and because it tends to be viewed in a rigid manner;

■ in the tables containing category descriptors

and examples, terms such as mild, moderate, severe, short-term and long-term are deliberately not defined further as interpretation will depend on the species being used, the details of its biology and the circumstances surrounding the manipulations involved. This underlines the importance of the tables and examples being used as indicative rather than definitive. Judgement must be exercised by the researchers and AEC members and this judgement must be informed by consultation with experts in the biology and behaviour of the particular species; and

■ a provisional score with respect to mental

state should be established first as the ultimate measure of impact. The impacts from the other four physical domains, as contributors to that ultimate measure, are then checked to ensure that no factor has been missed nor the impact with regard to mental state over- or under-estimated.

A number of other countries (e.g. Canada, Finland, The Netherlands, New Zealand, Switzerland and the United Kingdom) have also adopted the use of ‘harm scales’ as public policy. In these countries it is a mandatory requirement that investigators assess and record the severity of harm done to animals in biomedical research, typically according to categories of minor, moderate, or severe levels of invasiveness. Along with providing essential information to those involved in evaluating the justification of scientific procedures, the use of harm scales and other scoring systems for assessment of adverse states in laboratory animals promotes the application of the three R’s, with data from the Netherlands demonstrating a reduction in the severity of laboratory animal procedures (Orlans 2000). The use of severity scales can help to define clear upper limits on animal suffering which can assist in implementing humane end-points and can also identify procedures that cause the most animal suffering and target these as priorities for application of the three R’s (Smith and Jennings 2004).

A discussion group organised to consider the appropriateness and usefulness of the severity categorisation system of scientific procedures in the UK have come up with a number of suggestions that could be equally applicable to the assessment of severity of impact of pest animal control methods (Smith and Jennings 2004).The group suggested that a severity assessment should:

■ focus on the individual animal;

■ be assessed from the animal’s point of view

as far as possible; and

Common carp (Cyprinus carpio) (photo by John Gasparotto)

(24)

■ adopt a ‘holistic’ approach, in which there is

an attempt to consider all factors that can potentially influence well-being including psychological effects (e.g. anxiety, fear, boredom), physical effects, duration of effects as well as wider factors such as transport and husbandry.

The group also suggested that guidance on assessment of animal suffering and how to assign severity categories should:

■ cover all classes of vertebrates and

protected invertebrate species, as well as their protected developmental stages;

■ encompass a wide range of different kinds

of adverse effects (including their duration), protocols and techniques;

■ as far as possible be based on empirical

evidence; and

■ include detailed worked examples to illustrate

the application of severity categories in practice, particularly at their boundaries. The group noted that it is particularly difficult to assign severity categories when adverse effects are uncertain or unpredictable.

A 4c Assessment of production

animal welfare

A number of approaches to the assessment of animal welfare in production animals have been reported in the literature, with most taking an integrated approach. ‘Overall welfare assessment’ aims to assess welfare based on knowledge of the biological needs of animals and usually involves combining a number of weighted, welfare-relevant attributes or criteria to produce an overall welfare score (Bracke et al. 1999a). Scott et al. (2001) describe a methodological framework for the development of a composite animal welfare scale based on a number of individual welfare-related items. This involves the use of a scaling procedure to combine separate

items to create a single welfare measure. The technique follows psychometric and metrological principles for scale creation that were originally developed in the fields of human medicine and psychology. The stages in creating such a welfare assessment framework are:

1. Identification of the items to be included in the composite scale. These would be key components

of animal health and disease, behaviour and husbandry as well as more subjective factors which would help to assess the animal’s quality of life. The items to be included would be identified by surveying individuals involved in the farming area of interest (e.g. farmers, veterinary surgeons, animal welfare scientists). Once the items are listed, individuals are asked to rank the terms they associate with good welfare. The list is then reduced to a smaller list of items containing expressions relating to disease, management practices and behaviour.

2. Construction of a composite welfare index. A

scaling technique is used to allow for weighting of the items to reflect the level of welfare associated with them. Expert judgement (gathered from a large body of experts) would be used to assign the relative weights. After weighting, the individual items are combined to form a single composite measure.

3. Testing. The resulting composite index must be

validated and its reliability assessed by repeated use with multiple observers under a number of experimental conditions. Amendments to the draft welfare index may be necessary following this testing stage.

A similar, although more complex, approach has been used to construct a system for overall welfare assessment in pregnant sows (Bracke et al. 2002). This model is implemented in a computer-based decision support system that takes a description of a housing and management system as input and produces a welfare score as output. The welfare status of pregnant sows is assessed in relation to their housing and management

(25)

system based on available scientific knowledge. The model contains 37 attributes such as ‘space per pen’, ‘exposure to cold’, ‘handling and fear’, ‘resting comfort’ and ‘social stability’ that describe the welfare relevant properties of housing and management systems. These attributes are linked to statements of need and scientific statements about the various welfare performance criteria. Weighting factors that represent the relative importance of the attributes are derived from the scientific statements. The welfare score is calculated as the weighted average score and is expressed as a value between 0 and 10. The advantage of this system is that it quantifies pregnant-sow welfare using a systematic and transparent procedure that covers all reasoning steps from selection of attributes to the determination of overall welfare status. It also has the flexibility to incorporate new insights about welfare assessment when they become available (Bracke et al. 2002).

The animal needs index (ANI) is an instrument for assessing and grading livestock housing with respect to the well-being of the animals (Bartussek 1999). It considers five components of the animal’s environment: (1) the possibility of mobility; (2) social contact with members of the same species; (3) condition of the floors on which animals are lying, standing and walking; (4) stable climate (including ventilation, light and noise; and (5) the intensity of human care. Conditions that are considered to improve animal welfare are given more points and the overall sum of the points gives the ANI-value. The values have been graded into different categories of good or poor welfare. The ANI is used in actual policy decision making in Austria, mainly in controlling organic farming and in connection with animal welfare legislation (Bartussek 1999; Bracke et al. 1999b). For a detailed review of the overall assessment of farm animal welfare refer to Botreau et al. (2007a) and Botreau et al (2007b).

A 4d Assessment of welfare of

free-living wild animals

In considering the impact of human activities and human-induced environmental changes on the welfare of free-living wild animals, Kirkwood

et al. (1994) proposed that at the simplest level,

the scale and severity of harm can be evaluated by considering the following four factors:

1. The number of animals affected. 2. The cause and nature of harm.

3. The duration of harm.

4. The capacity of the animal to suffer. These parameters should then be used to produce a summary that allows comparisons between cases. The summary should include the following components:

1. A description of the cause.

2. A description of the effect, based on observations or inferred knowledge about the cause.

3. Judgement of the levels of stress and/or pain caused.

4. A description of the magnitude of the problem (based on the numbers affected and mean duration of harm).

The authors warn that the process of allocating a score to reflect the severity of harm to welfare should be used with great caution due to a number of difficulties with this approach. They maintain that compiling a summary that includes the four components described above would provide the most useful picture of welfare impact caused by human activities. With regard to animal suffering, the authors take the view that that although all mammals and birds have the capacity to suffer the unpleasant sensations of pain or stress, there is insufficient information to grade this suffering.

(26)

Jordan (2005) states that science-based welfare assessment in wild animals can be difficult because species react differently to pain, stress and fear. Since physiological examination is not possible in the wild, reliance must therefore be placed on a detailed knowledge of normal animal behaviour as well as situations that cause poor welfare.

With regard to the welfare of pest animals, Broom (1999) suggests that during evaluation of a pest control procedure, the extent of poor welfare can be multiplied by the duration of poor welfare to get an estimate of the severity of the problem. To evaluate the effects of killing methods on welfare two kinds of measurement are required. These are:

1. The severity of any poor welfare before death.

2. The duration of the period during which the poor welfare continues.

Broom (1999) advocates using a cost-benefit approach with the adverse effects of the pest being compared with the extent of poor welfare of the pest animals that a control method would cause.

Animal welfare research has not historically focused on pest animals or their management, and for many methods of pest control their impact on welfare is not known (Littin and Mellor 2005). A number of reviews of animal welfare issues arising from the use of pest animal control methods have suggested approaches for their assessment (e.g. Sainsbury et al. 1995; Gregory 2003; Littin et al. 2004; Mellor and Littin 2004; Littin and Mellor 2005). But whilst current guidelines for assessing humaneness tend to focus on leg-hold traps and poisons, there is a need to evaluate the welfare impact of a wider range of methods. The next section is a summary of some science-based comparisons of humaneness or acceptability conducted on a range of pest animal management methods.

A5. Application of welfare/

humaneness assessment:

some examples relevant

to pest animal control

A 5a Assessment of humaneness

of traps

The humaneness of restraining traps (i.e. leg-hold and cage traps) is most often assessed by identifying the physical trauma caused by the trap to the captured animal. A number of studies have used an injury scoring or rating system to quantify the extent of injury caused by the trap and to compare the severity of injuries caused by different types of trap (Kreeger et al. 1990; Meek et al. 1995; Hubert et al. 1996; Fleming

et al. 1998; Woodroffe et al. 2005). Some studies

have also documented the physiological (e.g. elevation of heart rate, body temperature, cortisol, muscle enzymes, bilirubin, neutrophils etc.) and/ or behavioural (e.g. changes in activity levels, digging, pacing, chewing on trap) responses to trapping (Jacobsen et al. 1978; Kreeger et al. 1990; White et al. 1991; Schutz et al. 2006). However, to date there is no objective scoring system for restraining traps that integrates physical injuries with behavioural and physiological responses, at least in part because interpreting such responses is not straightforward (Powell and Proulx 2003).

The humaneness of traps that are designed to kill an animal (kill traps) is usually evaluated on the basis of the time it takes for the trap to render an animal insensible to pain, most often measured by the loss of palpebral (blinking) reflex (Warburton et al. 2000). Many studies have used this criterion to assess the killing performance of traps and to determine if they are acceptably humane (Warburton and Hall 1995; Warburton and Orchard 1996; Warburton

et al. 2000; Warburton and Poutu 2002; Poutu

and Warburton 2003). It has been argued that setting the performance criteria for killing traps is easier than setting performance criteria for restraining traps, because time to insensibility

(27)

and death are relatively easy to define objectively compared with the injury, pain, anxiety and stress that may be experienced by restrained animals (Powell and Proulx 2003). In a review of trapping methods used in Europe and North America, the welfare performance of killing traps was evaluated using the additional criteria of likelihood of escape of injured animals, percentage of mis-strikes, trap selectivity, as well as time to unconsciousness (Iossa et al. 2007).

The International Organisation for Standardisation (ISO) has developed standards for the performance evaluation of traps for killing and restraining mammals (Warburton 1995; International Organisation for Standardisation (ISO) 1999a; 1999b; Harris et al. 2005). The ISO standards are considered to be the best currently available criteria for assessing the humaneness of restraining traps, although they have been criticised because they do not assess pain, physiological stress and long term-impact of some injuries, nor do they give guidelines as to how to avoid capture of non-target species (Harris et al. 2005). Another major criticism of the ISO standards is that the assessment of traps in an artificial setting is not likely to create the range of conditions and individual animal behaviour that is likely to occur in field situations. This could lead to traps failing in the field and poor welfare of trapped animals. In New Zealand, the National Animal Welfare Advisory Committee has produced guidelines for assessing the welfare of restraining and kill traps used for mammals based on the ISO standards (NAWAC 2000). The aim of the NAWAC guidelines is to standardise the testing of welfare performance of traps, to improve the efficiency and selectivity of traps and also to encourage the development of new and existing traps to make them more effective and to reduce the extent of injuries and animal suffering. Traps are tested and assigned to one of two welfare performance classes (A or B) or if they do not pass the criteria, they are failed.

To assess the welfare performance of restraining

trap systems the guidelines confine the

measurement of predicted clinical impact on the

well-being of a trapped animal to observations of physical trauma or injury received. Thirty-five descriptions of trauma type are graded from 1 = no identifiable trauma, through to 35 = death. Trauma type is also more broadly classified into four classes i.e. mild, moderate, moderately severe and severe. This system is used to classify the overall trauma class e.g. if an animal receives 1 x mild trauma it is classified overall as mild, if it receives 1 x moderate or 3 x mild traumas, it is classified overall as moderate, if it receives 1 x moderately severe trauma, 2 x moderate traumas or 5 x mild traumas etc. it is classified overall as moderately severe, and so on. The guidelines stipulate what proportion of trapped animals is allowed to have trauma exceeding certain categories for a trap to pass the performance test. For killing

traps, the time to loss of corneal reflexes is used

as the assessment criterion For a trap to pass the test, stipulated proportions of trapped animals must be rendered irreversibly unconscious within 3 minutes to be classified as welfare performance Class A; or within 5 minutes to be classified as welfare performance class B.

The NZ NAWAC guidelines do not attempt to use any measures of psychological and physiological distress because “insufficient information exists on what physiological parameters to measure and, for any one parameter, what levels could be considered as the minimum” (p1, NAWAC 2000). Annex A of the guidelines however does provide a description of the types of physical injuries that traps can inflict and attempts to predict how these injuries might bring about a negative impact on the welfare of the animal e.g.

“Major subcutaneous soft tissue maceration

or erosion – covers a large area of soft tissue, perhaps half or full width of a limb, and possibly the entire thickness of the soft tissue. This will cause immediate pain and dysfunction of the affected body part. The animal might use the affected limb during flight, although it is likely to favour the limb. It will cause restriction in movement which may particularly affect hunting by predators, but will heal well with scar formation.” (p18, NAWAC 2000).

(28)

There has been some criticism of the injury scoring of restraining traps because a quantitative injury score is not a direct measurement of an injury level (Engeman et al. 1997). It is argued that application of a scoring system requires decisions on several levels of increasing

abstraction from the actual physical injuries. Also, inconsistencies in scoring of injuries can occur between observers and there can also be different general perceptions of what levels of injuries are unacceptable and how frequently they can occur before a trap type is considered unacceptable. Although there are some disadvantages, the current scoring or rating systems used for the assessment of trap humaneness does provide a systematic and objective way of evaluating the physical trauma caused by trapping systems, and these should be continued to be used in future trap evaluation (Harris et al. 2005). However, there are many other factors that need to be considered if an overall humaneness assessment is going to be made. These include:

■ Restraint time – the extent of injuries and

distress experienced by an animal caught in a foothold trap (or any live trap) is also influenced by the length of time spent in the trap. Longer restraint time is also a major factor in the development of dehydration or exposure and may also cause stress by disrupting natural behaviour and motivational systems (Schutz et al. 2006).

■ Method of euthanasia – consideration must

also be given to the method of euthanasia that will be used to kill the trapped animal (Harris et al. 2005). The benefits of having a relatively humane trapping system to capture an animal are countered if the method subsequently used for killing it is relatively less humane.

■ Effects of exposure or dehydration – trapping

systems that provide shelter from adverse weather conditions and food/water are likely to be more humane than those that don’t.

■ Anxiety/fear/stress – physical injury and pain

will obviously have a negative effect on the animal, but so too will anxiety caused by confinement/restraint and physical exertion related to struggling (Marks et al. 2004). White

et al. (1991) found that although foxes caught

in a box traps and padded leg-hold traps had no physical injury, they still had evidence of a ‘classical stress response’ (indicated by, amongst other things, elevated levels of blood adrenocorticotropin and cortisol) compared to control foxes. This stress response was more dramatic in the leg-hold trap caught foxes. Psychogenic factors (e.g. fear, surprise) and differences in the intensity of exertion (e.g. pacing for box trapped foxes and digging for foothold trapped foxes) were thought to be responsible for the increased stress and for differences in response between trap methods.

■ Pain – some injuries may only receive a low

or medium injury score but are capable of causing severe pain (e.g. sternal fractures, rib fractures, permanent tooth fracture with exposure of pulp cavity).

■ Long-term impact of injuries –animals that

escape a trap may sustain damage/injuries that can have serious long-term effects on welfare e.g. tooth damage or claw loss may result in an inability to catch prey, leg injuries could cause limping that result in predation, mouth injuries may prevent eating.

A problem with the last three of these factors is that they are rather difficult to assess.

A 5b Assessment of humaneness

of poisons

Whilst the humaneness assessment of traps currently relies on measures of physical injury or time to insensibility, the assessment of humaneness of toxic agents uses a wider set of criteria that includes behavioural, biochemical and pathological indicators. In the UK, the Food and Environmental Protection Act 1985 requires that methods for

(29)

controlling Pests should be humane and that they must be assessed for humaneness before they are registered for use (Pesticide Safety Directorate 1997; 2001). A UK MAFF working group, established to provide criteria for assessing humaneness, concluded that pain, di

A model for assessing the relative humaneness of pest animal control methods. Second edition June 2011